publications

2026

1. Putting reasons back into reasoning: how genuine reasoning is inference-based and why neuro-symbolic NLI could achieve it

   Reto Gubelmann

   Frontiers in Artificial Intelligence, 2026

   Abs DOI

   Taking Leibniz’ ideal of a universal truth-calculating machine as a vantage point, this article provides a philosophically sound analysis of the concept of reasoning in NLP. It argues that reasoning always involves inference, which in turn requires being guided by reason relations. Based on this, the article argues that Symbolic NLP is unable to reason for epistemic reasons on the part of humans, that Neural NLP is likely unable to do so in principle, and that Neuro-Symbolic NLP is ideally set up to succeed where the two approaches in isolation failed, that is, to progress toward realizing Leibniz’ vision of a truth-calculating machine—to the extent to which this is possible. We conclude by providing a theoretical grounding for the latter claim in the philosophy of a contemporary rationalist, namely Robert Brandom.

2025

1. Sentence Smith: Formally Controllable Text Transformation and Its Application to Evaluation of Text Embedding Models

   Hongji Li, Andrianos Michail, Reto Gubelmann, and 2 more authors

   2025

   Abs DOI

   We propose the SENTENCESMITH framework that enables controlled and specified manipulation of text meaning. It consists of three main steps: 1. Parsing a sentence into a semantic graph, 2. Applying human-designed semantic manipulation rules, and 3. Generating text from the manipulated graph. A final filtering step (4.) ensures the validity of the applied transformation. To demonstrate the utility of SENTENCESMITH in an application study, we use it to generate hard negative pairs that challenge text embedding models. Since the controllable generation makes it possible to clearly isolate different types of semantic shifts, we can gain deeper insights into the specific strengths and weaknesses of widely used text embedding models, also addressing an issue in current benchmarking where linguistic phenomena remain opaque. Human validation confirms that the generations produced by SENTENCESMITH are highly accurate.

2. Assessing Reliability and Political Bias In LLMs’ Judgements of Formal and Material Inferences With Partisan Conclusions

   Reto Gubelmann and Ghassen Karray

   In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Vienna, Austria, 2025

   DOI
3. Die Instrumente Der Rechtspraxis

   Reto Gubelmann and Peter Hongler

   2025
4. Identifying Open-Texture in Regulations Using LLMs

   Clement Guitton, Reto Gubelmann, Ghassen Karray, and 2 more authors

   Artificial Intelligence and Law, 2025

   Abs DOI

   Abstract Open-texture—e.g. vague, ambiguous, under-specified, or abstract terms—in regulatory documents lead to inconsistent interpretation, and are an obstacle to the automatic processing of regulation by computers. Identifying which parts of a legal text fall under open-texture is therefore a necessary requirement to make progress in automating the law. In this paper, we propose that large language models (LLMs) might provide an effective way to automatically detect open-texture in legal texts. We first investigate the obstacles by situating open-texture in the broader literature, and we test the hypothesis using two different LLMs—the proprietary gpt-3.5-turbo and the open-source llama-2-70b-chat—for the task of identifying open-texture in the General Data Protection Regulation. We evaluate their performance by asking 12 annotators to assess their output. We find, overall, that gpt-3.5-turbo overperforms llama-2-70b-chat on F 1 -scores (0.84 vs 0.67), and its high F 1 -score could make it a suitable alternative, or complement, to using human annotators. We also test the sensitivity of the findings against four further LLMs combined with six different prompts, and replicate a finding that there is low agreement between annotators when it comes to the identification of open-texture. We conclude the article by discussing the subjectivity of open-texture, the lessons to draw when testing for open-texture, and the consequences of using LLMs in the legal domain.

5. The Shift from Logic to Dialectic in Argumentation Theory: Implications for Computational Argument Quality Assessment

   Rositsa V Ivanova and Reto Gubelmann

   In Proceedings of the 31st International Conference on Computational Linguistics, Abu Dhabi, UAE, 2025

   Abs

   In the field of computational argument quality assessment, logic and dialectic are essential dimensions used to measure the quality of argumentative texts. Both of them have found their way into the field due to their importance to argumentation theory. We trace the development of core logical concepts of validity and soundness from their first use in argumentation theory to their understanding in state-of-the-art research. We show how, in the course of this development, dialectical considerations have taken center stage, at the cost of the logical perspective. Then, we take a closer look at the quality dimensions used in the field of computational argument quality assessment. Based on an analysis of prior empirical work in this field, we show how methodological considerations from argument theory can benefit state-of-the-art methods in computational argument quality assessment. We propose an even clearer separation between the two quality dimensions not only in regards to their definitions, but also in regards to the granularity at which the argumentative text is being annotated and assessed.

2024

1. Exploring the Usefulness of Open and Proprietary LLMs in Argumentative Writing Support

   Reto Gubelmann, Michael Burkhard, Rositsa V. Ivanova, and 3 more authors

   In Artificial Intelligence in Education. Posters and Late Breaking Results, Workshops and Tutorials, Industry and Innovation Tracks, Practitioners, Doctoral Consortium and Blue Sky, Cham, 2024

   DOI
2. Pragmatic Norms Are All You Need – Why the Symbol Grounding Problem Does Not Apply to LLMs

   Reto Gubelmann

   In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, Miami, Florida, USA, 2024

   Abs

   Do LLMs fall prey to Harnad’s symbol grounding problem (SGP), as it has recently been claimed? We argue that this is not the case. Starting out with countering the arguments of Bender and Koller (2020), we trace the origins of the SGP to the computational theory of mind (CTM), and we show that it only arises with natural language when questionable theories of meaning are presupposed. We conclude by showing that it would apply to LLMs only if they were interpreted in the manner of how the CTM conceives the mind, i.e., by postulating that LLMs rely on a version of a language of thought, or by adopting said questionable theories of meaning; since neither option is rational, we conclude that the SGP does not apply to LLMs.

3. Large Language Models, Agency, and Why Speech Acts Are beyond Them (for Now)–a Kantian-Cum-Pragmatist Case

   Reto Gubelmann

   Philosophy & Technology, 2024

   DOI

2023

1. Computer Supported Argumentation Learning: Design of a Learning Scenario in Academic Writing by Means of a Conjecture Map:

   Michael Burkhard, Sabine Seufert, Reto Gubelmann, and 2 more authors

   In Proceedings of the 15th International Conference on Computer Supported Education, Prague, Czech Republic, 2023

   DOI
2. A Loosely Wittgensteinian Conception of the Linguistic Understanding of Large Language Models like Bert, Gpt-3, and Chatgpt

   Reto Gubelmann

   Grazer Philosophische Studien, 2023

   DOI
3. Capturing the Varieties of Natural Language Inference: A Systematic Survey of Existing Datasets and Two Novel Benchmarks

   Reto Gubelmann, Ioannis Katis, Christina Niklaus, and 1 more author

   Journal of Logic, Language and Information, 2023

   Abs DOI

   Abstract Transformer-based Pre-Trained Language Models currently dominate the field of Natural Language Inference (NLI). We first survey existing NLI datasets, and we systematize them according to the different kinds of logical inferences that are being distinguished. This shows two gaps in the current dataset landscape, which we propose to address with one dataset that has been developed in argumentative writing research as well as a new one building on syllogistic logic. Throughout, we also explore the promises of ChatGPT. Our results show that our new datasets do pose a challenge to existing methods and models, including ChatGPT, and that tackling this challenge via fine-tuning yields only partly satisfactory results.

4. When Truth Matters - Addressing Pragmatic Categories in Natural Language Inference (NLI) by Large Language Models (LLMs)

   Reto Gubelmann, Aikaterini-lida Kalouli, Christina Niklaus, and 1 more author

   In Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023), Toronto, Canada, 2023

   Abs DOI

   In this paper, we focus on the ability of large language models (LLMs) to accommodate different pragmatic sentence types, such as questions, commands, as well as sentence fragments for natural language inference (NLI). On the commonly used notion of logical inference, nothing can be inferred from a question, an order, or an incomprehensible sentence fragment. We find MNLI, arguably the most important NLI dataset, and hence models fine-tuned on this dataset, insensitive to this fact. Using a symbolic semantic parser, we develop and make publicly available, fine-tuning datasets designed specifically to address this issue, with promising results. We also make a first exploration of ChatGPT’s concept of entailment.

2022

1. On What it Means to Pay Your Fair Share: Towards Automatically Mapping Different Conceptions of Tax Justice in Legal Research Literature

   Reto Gubelmann, Peter Hongler, Elina Margadant, and 1 more author

   In Proceedings of the Natural Legal Language Processing Workshop 2022, 2022

   Abs

   In this article, we explore the potential and challenges of applying transformer-based pre-trained language models (PLMs) and statistical methods to a particularly challenging, yet highly important and largely uncharted domain: normative discussions in tax law research. On our conviction, the role of NLP in this essentially contested territory is to make explicit implicit normative assumptions, and to foster debates across ideological divides. To this goal, we propose the first steps towards a method that automatically labels normative statements in tax law research, and that suggests the normative background of these statements. Our results are encouraging, but it is clear that there is still room for improvement.

2. A Phi\-losophically-Informed Contribution to the Generalization Problem of Neural Natural Language Inference: Shallow Heuristics, Bias, and the Varieties of Inference

   Reto Gubelmann, Christina Niklaus, and Siegfried Handschuh

   In Proceedings of the 3rd Natural Logic Meets Machine Learning Workshop (NALOMA III), 2022
3. Context Matters: A Pragmatic Study of PLMs’ Negation Understanding

   Reto Gubelmann and Siegfried Handschuh

   In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2022

   DOI
4. How NLP Can Improve the Quality of Arguments in Legal Studies

   Reto Gubelmann and Peter Hongler

   Allgemeine Juristische Praxis, 2022
5. Mechanism Versus Organism: A Loosely Kantian Perspective and Its Implications for Bioengineering

   Reto Gubelmann and Marco Toscano

   In Thinking: Bioengineering of Science and Art, 2022

   Abs DOI

   Building on the work of Immanuel Kant and researchersResearcher working in his tradition (but without getting lost in exegetical details that are immaterial for this volumeVolume), we discuss a notion of the living organismOrganismorganism, living based on its formal teleological structureStructurestructure, teleological, contrast it with mechanistic objectsObjectobject, mechanistic, and draw some consequences for the notion of bioengineering. Bioengineering, in our understanding, is close to a contradictio in adiecto: either something is alive, which implies substantial autonomyAutonomy in its inner functional organizationOrganizationorganization, functional, or it is engineered, which means that the engineer continuously determines the thing’s functional organizationOrganizationorganization, functional.

2019

1. A Science-Based Critique of Epistemological Naturalism

   Reto Gubelmann

   2019

   DOI
2. From Shared Stimuli to Preestablished Harmony: The Development of Quine’s Thinking on Intersubjectivity and Objective Validity

   Reto Gubelmann

   HOPOS – The Journal of the International Society for the History of Philosophy of Science, 2019

   DOI
3. Maddy vs. Quine on Innate Concepts. Revisiting a Perennial Debate in Light of Recent Empirical Results.

   Reto Gubelmann

   Philosophia, 2019

   DOI

2014

1. On Proving God’s Existence by Reason Alone. A Logical and Epistemological Study of Saint Anselm’s and Plantinga’s Ontological Arguments

   Reto Gubelmann

   2014